Plan
This deployment follows a two-phase plan to methodically build and test the solution.
About Core Pipeline Validation
The initial phase focuses on verifying the fundamental components of
the solution. You deploy and run the Morpheus GNN pipeline using static files for input
and output. This confirms that the Morpheus container, NVIDIA GPU drivers, Conda environment, and the AI models are all functioning correctly in an isolated
setting.
About Real-Time Streaming Integration
The second phase transitions the solution workflow to a real-time
model. This involves deploying an Apache Kafka service and modifying the Morpheus pipeline to use Kafka topics for ingesting data and publishing the results. This phase demonstrates the
solution's end-to-end streaming capability and its readiness for live data.